Skip to content
MF

Studio. Pilot.

Mercantile

Mercantile is built on a simple inversion. The best offer engine should not start from demand. It should start from commercial pressure.

Thesis.

Merchants do not have an offer problem in the abstract. They have a pressure problem. Inventory accumulates in the wrong places. Some products are over-discounted while others should never be touched. Slow stock hides inside healthy-looking catalogs. Trend spikes tempt merchants into broad promotions that lift revenue while quietly eroding gross profit. Today this gets solved with guesswork, spreadsheet analysis, discount apps, bundle apps, agencies, and consultants. Each piece solves a fragment. Nothing operates like a unified commercial decision engine.

Mercantile changes the unit of value. It does not produce coupons. It produces executable offer theses. Structured commercial moves assembled from pricing, bundling, timing, eligibility, inventory state, and margin constraints. The platform monitors the store, identifies where pressure is building, designs profit-safe offer paths, simulates likely outcomes, and deploys approved strategies into the storefront and checkout.

The category language is Retail Pressure Intelligence. The discipline of sensing commercial pressure inside a store and converting that pressure into profitable, executable offer strategies. Instead of asking what promotion could increase conversion, the system asks where commercial pressure is accumulating and what is the safest, smartest way to release it.

The hidden architecture.

Mercantile introduces its own vocabulary because category-defining products travel on named concepts. The Pressure Map is a live model of where commercial risk or opportunity is building across the catalog and the customer base. Every SKU, collection, and buyer segment receives a dynamic pressure profile across dimensions like slow sell-through, overstock, expiry risk, high subsidy capacity, anchor strength, reactivation potential, and cannibalization risk. The Offer Assembly is a structured commercial move composed from multiple levers rather than a single discount. Each generated Assembly comes with a confidence-weighted projection of revenue lift, gross profit impact, margin exposure, cannibalization risk, brand conditioning risk, and a defined rollback condition.

The system operates in four behavioral modes. Scout, which scans for pressure and hidden risk. Composer, which designs the offer assemblies. Governor, which applies financial and strategic constraints, including margin floors, customer eligibility, and saturation limits. Operator, which deploys, measures, adjusts, and rolls back. Each mode is explainable. The product does not feel like a dashboard asking the merchant to interpret charts. It feels like a commercial operator that explains itself.

Long-term.

Mercantile is in Shopify-native pilot now. The first version pairs a Shopify app for native access to products, orders, customers, inventory, and discount execution, with a SaaS control layer that acts as the commercial brain. Once the engine proves itself on Shopify, the expansion path is ERP, WMS, and POS integrations, wholesale and offline commercial plays, autonomous policy-bounded operation for selected accounts, and portfolio-level decision support for multi-brand operators.

The moat is not the AI. The moat is the Outcome Graph: a proprietary dataset that links store state, offer configuration, segment conditions, and downstream economics across thousands of campaigns. Every campaign teaches the system more about which configurations work, fail, distort behavior, or preserve margin under different store conditions. The longer Mercantile runs, the harder it becomes to replicate.